Curated Links: This Week's AI Stories

This week brought a convergence of stories that signal a fundamental shift in how AI is deployed across industries. Humanoid robots entered government spaces, open standards began dismantling software lock-in, and enterprise platforms started rate-limiting AI agents over security concerns. These are not isolated incidents. They represent the collision of three forces: autonomous systems going mainstream, software moats collapsing, and enterprise security catching up to innovation. Here are 20 stories that define this moment.

Figure 03 Enters the White House - Humanoid Robots Go Mainstream: First humanoid in the White House; McDonald's pilots robot staff in Shanghai; Amazon acquires home robotics startup Fauna. The deployment of humanoids in public-facing roles signals that autonomous systems are moving beyond controlled environments.

Construction Humanoids Summit Announced: Dedicated event exploring onsite application of robotics in the built environment. Governance frameworks must be in place before humanoids reach construction sites.

Wikipedia Bans AI-Generated Articles (40-2 Vote): AI allowed for grammar and translation only. Signals growing quality concerns around AI-generated content. The decision reflects a broader industry concern: as AI becomes easier to deploy, quality gatekeeping becomes harder to maintain.

Sakana AI's AI Scientist Published in Nature: Automated pipeline invents research ideas, runs experiments, writes papers, and passed peer review at a top ML conference. The contrast is stark: Wikipedia rejects AI-generated content while peer-reviewed research now accepts it.

Automated Internet Traffic Now Growing 8x Faster Than Human Traffic: OpenAI generating approximately 69 percent of all AI bot traffic; agentic AI traffic up 7,851 percent. The volume of autonomous AI agents operating on the internet now outpaces human activity.

Slack, Workday & LinkedIn Cracking Down on AI Agents: Major platforms rate-limiting or blocking external AI agent access over security concerns. Enterprise platforms are beginning to restrict agent access, signalling that security concerns are outpacing adoption enthusiasm.

LiteLLM Supply Chain Breach Hits 3.4M Downloads: Open-source AI tool infected with credential-stealing malware. The vulnerability highlights the risks of rapid AI tool adoption without proper security vetting.

Anthropic Considering October IPO at $60B+: Could be second-largest tech IPO ever behind SpaceX. The valuation signals market confidence in enterprise AI adoption and Anthropic's focus on safety.

Anthropic's AI Skills Gap Research: Power Users Pulling Ahead: 6+ month users have 10 percent higher success rate. The difference is not AI capability. It's human discipline in how we deploy AI.

Novo Nordisk Deploys AI Agents in Clinical Trials: Ozempic maker says agents are shortening approval timelines by weeks or months. Enterprise AI agents are moving beyond experimentation into measurable business impact.

Why Your People Are Resisting Your AI Rollout: Resistance to AI adoption is not about the technology. It's about how change is managed and communicated. Understanding this gap is critical for enterprise rollouts.

Meta Launches New Initiative to Support Entrepreneurship, Drive AI Adoption: Meta is backing startups building on its AI infrastructure. The move signals confidence in the AI startup ecosystem and accelerates adoption across sectors.

ChatGPT Ads Are Now Live: OpenAI hired Meta's former ad VP; $200k minimum spend. The commercialisation of AI assistants is fundamentally changing how they are monetised.

Gemini 3.1 Flash Live Goes Global: Real-time voice AI now in 200+ countries, 90+ languages. Deployed by Verizon and Home Depot. Enterprise adoption will follow consumer familiarity.

Apple Opens Siri to Rival AI Assistants in iOS 27: Users can choose their preferred AI model via Extensions settings. Apple's decision to open Siri signals a shift in how AI assistants will be distributed into enterprise workflows.

Microsoft Is Removing 'Unnecessary' Copilot Features in Windows: Microsoft is streamlining Copilot after user feedback. The move reflects a broader trend of companies refining AI features based on real-world usage patterns.

OpenAI 'Superapp' to Merge ChatGPT, Codex, and Atlas Browser: OpenAI is consolidating its tools into a single platform. The move signals a shift toward integrated AI experiences rather than standalone tools.

Perplexity Can Now Access Your Apple Health Data to Answer Medical Questions: Perplexity gains access to personal health data for contextual AI responses. The integration raises questions about data privacy and AI access to sensitive information.

Bret Taylor's Sierra Launches Ghostwriter - An AI Agent That Builds AI Agents: Companies can spin up customer service bots across voice, chat, and 30+ languages. The abstraction layer for building agents is lowering rapidly.

Dan Shipper's Agent-Native Architecture Framework: AI agents are the product core, shipping with virtually zero hand-written code. Traditional software development workflows are becoming obsolete.

He Rebuilt 6 Years of Work in 4 Days with AI: Brian Alvey (WordPress VIP) shares how AI collapsed his project timeline. This is evidence that AI is not incrementally improving productivity. It's collapsing timelines for specific tasks.

Bricks & Bytes: Trayd Closes $10M Series A in 3 Weeks for Construction Tech: Transforming back-office software for contractors. The velocity of funding signals that investors see genuine opportunity in AI-driven back-office automation.

5 Things Learned from a Month of Executive Briefings: Bricks & Bytes spoke to CTOs, PhDs, VPs, attorneys, and AI founders. This synthesis provides a snapshot of how construction leaders are thinking about AI adoption.

Promise of AI-Driven Efficiencies Draws Investors to Hotels, C&W Reports: AI-driven operational improvements are attracting capital to the hospitality sector. The trend shows AI adoption is expanding beyond tech into traditional industries.

Are You Optimising for the Wrong Thing? - Pablo Zamorano on AI and Design Values: AI makes mediocrity free, firms need ideas worth realising. Language-based image generation became their most adopted tool in month one.

OpenAI Kills Sora Video Platform, Preps Spud Model: Video generation shuttered, GPUs redirected to new model; Disney $1B deal voided. OpenAI's decision signals a strategic pivot away from consumer video tools.

That's this week's curated selection. These stories represent the frontiers of AI adoption, from enterprise security to autonomous systems to the fundamental reshaping of how software is built and deployed.

All content reflects our personal views and is not intended as professional advice or to represent any organisation.